Myeloperoxidase formation of PAF receptor ligands induces PAF receptor-dependent kidney injury during ethanol consumption

Abstract

Cytochrome P450 2E1 (CYP2E1) induction and oxidative metabolism of ethanol in hepatocytes inflame and damage liver. Chronic ethanol ingestion also induces kidney dysfunction, which is associated with mortality from alcoholic hepatitis. Whether the kidney is directly affected by ethanol or is secondary to liver damage is not established. We found that CYP2E1 was induced in kidney tubules of mice chronically ingesting a modified Lieber-deCarli liquid ethanol diet. Phospholipids of kidney tubules were oxidized and fragmented in ethanol-fed mice with accumulation of azelaoyl phosphatidylcholine (Az-PC), a nonbiosynthetic product formed only by oxidative truncation of polyunsaturated phosphatidylcholine. Az-PC stimulates the inflammatory PAF receptor (PTAFR) abundantly expressed by neutrophils and kidney tubules, and inflammatory cells and myeloperoxidase-containing neutrophils accumulated in the kidneys of ethanol-fed mice after significant hysteresis. Decreased kidney filtration and induction of the acute kidney injury biomarker KIM-1 in tubules temporally correlated with leukocyte infiltration. Genetic ablation of PTAFR reduced accumulation of PTAFR ligands and reduced leukocyte infiltration into kidneys. Loss of this receptor in PTAFR(-/-) mice also suppressed oxidative damage and kidney dysfunction without affecting CYP2E1 induction. Neutrophilic inflammation was responsible for ethanol-induced kidney damage, because loss of neutrophil myeloperoxidase in MPO(-/-) mice was similarly protective. We conclude that ethanol catabolism in renal tubules results in a self-perpetuating cycle of CYP2E1 induction, local PTAFR ligand formation, and neutrophil infiltration and activation that leads to myeloperoxidase-dependent oxidation and damage to kidney function. Hepatocytes do not express PTAFR, so this oxidative cycle is a local response to ethanol catabolism in the kidney.

Keywords: Acute kidney injury; Free radicals; Inflammation; Kidney; Oxidized phospholipids; PAF receptor; Reactive oxygen species.

Figure 1. Linoleoyl sn-2 residues of ether phosphocholines are truncated by radical oxidation

Hydrogen ion extraction between vicinal ethanolic bonds of polyunsaturated fatty acyl residues is followed by rearrangement to direct molecular oxygen incorporation at the distal and proximal end of the run of double bonds, forming phospholipid hydroperoxide (hydro)peroxides at these sites. The abundant sn-2 linoleoyl residue thereby produces nine carbon azelaoyl (nonanedioic acid) residues that remain esterified to the glyceryl phospholipid backbone. Oxidative truncation of linoleyl choline phospholipid to azelaoyl choline phospholipid generates a PAF-like agonist that stimulates the G protein coupled receptor for Platelet-activating Factor (PAF), the PTAFR. The critical elements recognized by this G protein coupled receptor include the short sn-2 acyl chain and the choline sn-3 headgroup;

Figure 2. Chronic ethanol ingestion alters renal architecture and function

A) Chronic ethanol ingestion alters the structure of murine kidney. Paraffin embedded kidney sections were deparaffinized and hydrated, stained with Periodic acid-Schiff reagent (60x). Lower Expanded images of glomeruli (G, left) or proximal tubules (PT, right) of mice pair-fed a control diet (Left) or 25d ethanol (Right) as described in “Methods”. Ethanol-fed mice kidneys have swollen tubules with proteinaceous substances within the tubules, and mesangial matrix expansion around glomeruli. B) Ethanol feeding functionally damages kidney. Blood urea nitrogen (BUN) and creatinine (mg/dL) in plasma were increased in ethanol-fed mice relative to control mice. The data are expressed as mean±SEM (n=6) and p<0.05 was considered as significant. C) Ethanol feeding induced mRNA for Kidney Injury Molecule-1. Total RNA was isolated from ethanol and pair-fed kidneys at the stated times during the feeding trial, and quantified by SYBR Green one-step Reverse Transcription-PCR for CYP2E1 and 18S with the Bio-Rad MyiQ real-time PCR detection system. mRNA expression was normalized to 18S mRNA content and 2^−ΔΔCT was used to calculate the fold changes. Data are expressed as mean±SEM (n=4), and p<0.05 was considered as significant.

Figure 3. Chronic ethanol feeding induces renal CYP2E1 and local oxidative stress

A) Immunohistochemistry of CYP2E1 expression in murine kidney tubules. Paraffin embedded kidney sections of animals fed ethanol or pair fed an isocaloric control diet for 25 days were deparaffinized, hydrated, and blocked with 10% donkey serum. The sections were immunostained with anti-rabbit CYP2E1 antibody, and Alexa Fluor^® 488 donkey anti-rabbit IgG as a secondary antibody prior to imaging fluorescence at 60X. The figures are representative of 2 to 3 images per kidney of 4 mice per group. B) Ethanol induces CYP2E1 protein expression. Western blot of CYP2E1 expression in the crude homogenate (20 μg protein) of kidneys from pair-fed and ethanol-fed mice detected with IRDye^® 800 CW anti-rabbit secondary antibody prior to scanned in an Odyssey Image station. Equal loading of protein was confirmed by using anti-β-actin antibody and Li-Cor IRDye® 680 RD goat anti-mouse IgG₁ secondary antibody. The box defines the fold change in gray scale imaging ratios. C) Ethanol feeding induces CYP2E1 mRNA accumulation. Total RNA was isolated from ethanol and pair-fed kidneys, and quantified by SYBR Green one-step Reverse Transcription-PCR for CYP2E1 and 18S by real-time PCR. The mRNA expression was normalized to 18S mRNA content and 2^−ΔΔCT was used to calculate the fold changes. The data are expressed as mean±SEM (n=4), and p<0.05 was considered as significant. D) Ethanol ingestion induces 4-hydroxynonenal adduct formation in renal tubules. Paraffin embedded kidney sections were deparaffinized and hydrated, antigens retrieved as described in “Methods.” Peroxidase was blocked, the sections blocked with blocking buffer, and then stained with anti-4-HNE antibody followed by HRP-conjugated secondary antibody that was developed with DAB/metal staining. Dark brown staining identifies 4-hydroxynonenal protein-adducts. The figures are representative of 2 to 3 images per kidney of 4 mice per group. E) CYP2E1 expression in transiently transfected human kidney cells. HK-2 cells were transfected with CYP2E1 (pCI-2E1) and Lipofectamine for 48 h, harvested and reseeded for 24h in 10% FBS/DMEM. Cells were treated, or not, with ethanol (100 mM daily) for 4 days prior to harvest and solubilization for SDS-PAGE immunohistochemistry with anti-CYP2E1 antibody followed by anti-β-actin antibody. F) CYP2E1 expressing HK-2 cells accumulate PTAFR ligands. HK-2 cells transiently transfected with CYP2E1, or not, were exposed to ethanol as in the preceding panel, although with or without the CYP2E1 inhibitor diallyl sulfide (100 μM). Total cellular lipids were extracted, dried, and reconstituted in HBSS with 0.5% human serum albumin before the presence of PTAFR ligands was determined using freshly isolated human neutrophils previously loaded with FURA2-AM. Changes in intracellular free Ca⁺⁺ was monitored by changes over time in the fluorescent intensity at 340 nm relative to 380 nm. Synthetic PAF (1 nM) was the positive control. The table is a compilation of results using extracts of HK-2 cells treated as described where the fluorescent ratio of each assay is presented before and after extract addition.

Figure 4. Chronic ethanol ingestion induces phospholipid oxidation, formation of PAF receptor ligands, and leukocyte infiltration into murine kidney

A) Oxidized phospholipid adducts form in kidney tubules during ethanol ingestion. Paraffin embedded kidney sections were deparaffinized, hydrated, and subjected to antigen retrieval before blocking with 10% goat serum as in the proceeding panel. The processed sections were incubated with anti-oxidized phospholipid E06 antibody (1:1000) before detection with Alexa Fluor^® 568 goat anti-mouse IgG (1:1000). The panels (60X) are representative of 2–3 images per kidney of 4 mice per group. B) PTAFR ligands accumulate in kidney after ethanol feeding. Total lipid was isolated from pair-fed and ethanol-fed kidneys in the presence of [²H]PAF as an internal standard. sn-2 Palmitoyl azelaoyl choline phospholipid (Az-PC) and Platelet-activating Factor (PAF) were quantified by liquid chromatography/electrospray ionization/tandem mass spectrometry (LC/MS/MS) as described in methods and normalized to the amount of protein in the extracted tissue. The data are expressed as mean±SEM (n=4), and p<0.05 was considered as significant. C) mRNA for leukocyte β₂ integrin (CD18) increased in kidney during ethanol ingestion. Mice were pair-fed an isocaloric control diet that was substituted with progressively increasing concentrations (v/v) of ethanol for the stated times. Total RNA was isolated from pair-fed and ethanol fed mice kidneys, and quantified by SYBR Green one-step reverse transcription-PCR for CD18 and 18S by real-time PCR detection. mRNA expression was normalized to 18S mRNA content and 2^−ΔΔCT was used to calculate the fold change. The data are expressed as mean±SEM (n=4), and p<0.05 was considered as significant. D) Ethanol feeding increases mRNA for myeloperoxidase in kidney over time. RNA was extracted and processes as in the preceding panel before quantitation by real time qPCR relative to 18S RNA. The data again expressed as mean±SEM (n=4), and with significance set at p<0.05.

Figure 5. PTAFR is required for ethanol-induced kidney tubule oxidation and PTAFR ligand formation

A) CYP2E1 induction is independent of PTAFR. Kidney homogenates from wild-type and PTAFR^−/− mice fed control or ethanol-containing diets were resolved by SDS-PAGE (20 μg protein/well) prior to immunostaining for CYP2E1 with anti-rabbit CYP2E1 antibody followed by IRDye^® 800 CW anti-rabbit secondary antibody prior to scanned in an Odyssey Imaging system. We again confirmed equal loading of protein simultaneously using anti-β-actin antibody and IRDye® 680 RD goat anti-mouse IgG₁ secondary antibody. Fold change from pair fed control homogenates was calculated (box) following conversion to gray scale. B) 4-hydroxynonenal adduct formation in renal tubules requires PTAFR. Sections of kidney of C57/BL6 or PTAR^−/− mice fed control or ethanol supplemented diets were immunostained with anti-4-hydroxynonenal antibody as in Fig. 3. The dark brown deposits of protein-4-hydroxynonenal adducts are representative of 2 to 3 images per kidney of 4 mice/group. C) PTAFR ligand formation in the kidneys of ethanol fed mice requires PTAFR. sn-2 Azelaoyl choline phospholipid (Az-PC) and Platelet-activating Factor (PAF) were quantified in kidneys of pair- and ethanol-fed wild-type and PTAFR^−/− mice by liquid chromatography/electrospray ionization/tandem mass spectrometry (LC/MS/MS) using [²H]PAF as an internal standard as in the preceding figure. The data are expressed as mean±SEM (n=4), and p<0.05 was considered as significant.

Figure 6. Genetic ablation of PTAFR attenuates ethanol-induced neutrophilic inflammation and protects kidney function

A) Genetic deletion of PTAFR blocks ethanol-induced renal inflammation. Parental C57BL/6 mice and their PTAFR^−/− progeny were pair-fed a control diet or the progressive ethanol diet before total RNA was isolated from pair-fed and ethanol fed mice kidneys, and mRNA was quantified by real time reverse transcription-PCR for CD18, CD64, MPO. mRNA expression was normalized to 18S mRNA content and 2^−ΔΔCT was used to calculate the fold change as before. The data are expressed as mean±SEM (n=4), and p<0.05 was considered as significant. B, C) PTAFR deletion prevents kidney inflammation and damage. mRNA for the inflammatory mediator B) TNFα and C) for the acute kidney injury marker KIM-1 was quantified and analyzed as in the preceding panel. D) Ablation of the PAF receptor suppresses KIM-1 protein expression in kidney tubules. Paraffin embedded kidney sections were processed as in Fig. 3 prior to immunostaining for KIM-1 (1:1000,) with detection by Alexa Fluor 488 donkey anti-rabbit IgG as a secondary antibody. The kidney sections were observed under fluorescence microscopy. The figures (60X) are representative of 2–3 images per kidney of 4 mice per group. E) Kidney function is protected by PTAFR deletion. Plasma creatinine and blood urea nitrogen (BUN) concentrations (mg/dL) were assessed by commercial kits as in Fig. 2. Data are expressed as mean±SEM (n=4), with significance accepted at p<0.05.

Figure 7. Genetic deletion of myeloperoxidase attenuates ethanol-induced kidney damage and dysfunction

A) Kidney structure is preserved in MPO^−/− mice ingesting ethanol. Kidney from C57BL/6 parental mice or their MPO^−/− derivatives were sectioned, fixed, and processed as in Fig. 2 prior to staining with Periodic acid–Schiff (PAS) and hematoxylin reagent and imaging at 60x (Top). Glomeruli (G, middle) or proximal tubules (PT, lower) from each section imaged at 60x were electronically expanded. B) Genetic ablation of MPO blocked ethanol induced acute kidney injury. Kidneys of wild-type or MPO^−/− mice ingesting ethanol or pair fed control diets were harvested before KIM-1 immunofluorescence was quantified in 4 to 5 sections in each kidneys of pair-fed and ethanol-fed mice using ImageJ 1.47v software (NIH), and the data are expressed as mean±SEM, n=4, p<0.05 (*). C) MPO deletion protects renal filtration from ethanol ingestion. Plasma creatinine and blood urea nitrogen (BUN) concentrations (mg/dL) were assessed as in Fig. 2. The data are expressed as mean±SEM (n=4), and p<0.05 was considered as significant.

Fig. 8

Ethanol metabolism induces a cycle of PTAFR ligand formation followed by PTAFR-dependent activation of neutrophil myeloperoxidase. CYP2E1 is induced in kidney tubules by ethanol allowing CYP2E1 production of ethanol to create reactive oxygen species (ROS) that peroxidize cellular polyunsaturated phospholipids. Hock fragmentation generates reactive fragments that derivatize local proteins so local tissue oxidation is marked by anti-4-hydroxynonenal and E06 anti-phospholipid antibody. Oxidative fragmentation of phosphatidylcholine also generates PAF-like truncated phospholipids that stimulate the PAF receptor (PTAFR). PTAFR stimulates neutrophil infiltration and activation that both synthesize PAF and releases oxidants that contribute to oxidative phospholipid truncation to PTAFR ligands. PTAFR activated neutrophils release myeloperoxidase (MPO) that damages kidney function. Genetic ablation of PTAFR blocks initiation of this new cascade, while ablation of myeloperoxidase blocks tissue damage and PTAFR ligand formation.